Ready to see CodeRabbit in action? Try our Quickstart
guide for a hands-on demonstration using a real
repository.
What CodeRabbit does for your pull requests
CodeRabbit automatically analyzes every pull request with a multi-layered approach that combines the best of AI and industry-standard tools:AI-generated summaries
Comprehensive summaries and walkthroughs of code changes with contextual
insights
Security & quality analysis
Integration with 40+ open-source linters and security scanners for
comprehensive coverage
Intelligent suggestions
Context-aware improvement recommendations based on your entire repository
Code graph analysis
Deep understanding of code relationships and dependencies across your
project
How automatic reviews work
1
Integration setup
After you connect CodeRabbit to your repository, it monitors
for new pull requests and commits
2
Instant analysis
When a pull request is created, CodeRabbit immediately begins analyzing the
code changes using multiple AI models and static analysis tools
3
Comprehensive review
Within minutes, CodeRabbit publishes detailed review comments with
summaries, security findings, and improvement suggestions
4
Continuous updates
For subsequent commits, CodeRabbit performs incremental reviews focusing on
the new changes
CodeRabbit’s analysis includes Code Graph
Analysis when available, providing deeper
insights into how your changes affect the broader codebase structure.
Review types and severity levels
CodeRabbit categorizes its feedback into different types and severity levels to help you prioritize and address issues effectively.Review types
CodeRabbit provides three types of review feedback:- ⚠️ Potential issue - Identifies potential bugs, security vulnerabilities, or problematic code patterns
- 🛠️ Refactor suggestion - Recommends code improvements for maintainability, performance, or best practices
- 🧹 Nitpick - Suggests minor style or formatting improvements (only in Assertive mode)
Severity levels
Each review comment is assigned a severity level to indicate its importance:- 🔴 Critical - Severe issues that could cause system failures, security breaches, or data loss
- 🟠 Major - Significant problems that impact functionality or performance
- 🟡 Minor - Issues that should be addressed but don’t critically impact the system
- 🔵 Trivial - Low-impact suggestions for code quality improvements
- ⚪ Info - Informational comments or context without requiring action
Review triggers and events
CodeRabbit automatically initiates reviews based on these repository activities:- New pull requests
- New commits
Full comprehensive review when a new pull request is created - Complete
analysis of all proposed changes - Security and quality assessment - Code
style and best practices review
Interactive code reviews with CodeRabbit
Once CodeRabbit reviews your pull request, you can engage in dynamic conversations and request specific actions by mentioning@coderabbitai in your comments.
Chat & discuss
Have natural conversations about your code with full repository context
Control reviews
Pause, resume, or customize review behavior with simple commands
Generate code
Request CodeRabbit to implement suggestions and create improvement PRs
Smart conversation capabilities
- Contextual chat
- Review control
- Code generation
Ask CodeRabbit questions about your code changes, architecture decisions, or implementation approaches. It has access to your entire repository for informed responses.
CodeRabbit learns from your feedback and coding patterns to provide
increasingly relevant suggestions over time.
Next steps
Ready to dive deeper into CodeRabbit’s capabilities? Explore these essential features to maximize your code review experience:Review commands
Learn all the commands to control CodeRabbit’s behavior during reviews
Generate improvements
Let CodeRabbit create pull requests with code improvements for you
Agent chat
Have in-depth conversations about your code with AI assistance
Code review best practices
Tips and strategies for effective code reviews with CodeRabbit